Safety of Test Result Follow-Up: Challenges, Progress, and More Challenges

• 5/15/2010
• Author: Steven Berman
• Category: The Journal Blog
• 10723 Views
• 0 Comments
• 

This week, Hardeep Singh and Meena S. Vij, the authors of “Eight Recommendations for Policies for Communicating Abnormal Test Results,”  which appeared in the May 2010 issue, are guests on the Journal blog. The evidence- and experience-based recommendations provided in the article are intended to help organizations ensure safe and timely abnormal test-result communication. The article was accompanied by an editorial, “Eight Questions for Getting Beyond ‘Getting Results,’” by Gordon D. Schiff, to which Dr. Singh now responds. We welcome your comments.

We thank Dr. Schiff for commenting on our recommendations for policies for better test-result communication and for envisioning the next steps for this work. As he points out, improving test-result communication is often presumed to be an easy fix to prevent diagnostic errors. In reality, as we both have described, this “low-hanging fruit” continues to elude us.

Implementation of new policies and monitoring communication outcomes and processes over time, are essential to advancing this field. Thus, Questions 1–3 and 6 are integral to the portfolio of our future work. As Dr. Schiff suggests, we now have baseline data and reliable methods for measuring outcomes. However, we approach longitudinal assessment with a measure of caution. Identifying which improvements, if any, were related to our policy itself is an ongoing challenge within a system that is subject to many other internal and external influences at both local and national levels.

Regarding Question 4, we fully agree that there is potential to overwhelm busy clinicians with test-result notifications. This is precisely the reason that the types of subcritical test results that Dr. Schiff discusses are not always communicated verbally to clinicians. Our own policy provides one example of how to prioritize handling of critical and abnormal findings to address the “signal-to-noise ratio” problem. Meanwhile, we agree that it is also important to carefully consider the timing and mode of test-result delivery to patients themselves (Question 7). We believe that the Department of Veterans Affairs (VA) has already taken the lead by initiating the VHA Directive mentioned in the article; specific patient-centered language is a part of this document.

Whom to notify and how (Question 5), are complex questions that our policy also attempts to address. Currently, we are unaware of any other institutions in the United States that have similarly detailed protocols in place. Our policy not only provides general guidance for patient ownership to providers who “just ordered the test” but also sets an example for others to follow by elaborating detailed protocols in the Appendix. Although we do not claim to have an optimal solution, we believe that raising the concern about test-result follow-up responsibility is a necessary start. For instance, in the example provided by Dr. Schiff, the surgical consultant might object to following up on the renal mass on the MRI but may be more likely to accept the responsibility for personally notifying the patient’s primary care physician to ensure follow-up.  We encourage others to develop clear, detailed safeguards pending the development of truly “fail-safe” systems.  

Finally, we agree that our ultimate goals lie beyond merely “getting” test results (Question 8). As we have seen in our previous work, test results are missed even in systems that virtually guarantee their delivery to providers. Thus, one apparent “fix” in the communication process exposes problems elsewhere in the test-result life cycle. While having effective policies and protocols for communication are a start, questions about the “newly” discovered complicated problems are now informing our research agenda. Indeed, the road ahead is not so smooth as we once thought, and it is sure to keep us in this line of work for years to come.